Breast cancer survivors’ supportive care needs 2–10 years after diagnosis
- First Online:
- Cite this article as:
- Hodgkinson, K., Butow, P., Hunt, G.E. et al. Support Care Cancer (2007) 15: 515. doi:10.1007/s00520-006-0170-2
- 1.2k Views
Goals of the work
A significant proportion of breast cancer patients experience psychosocial morbidity after treatment, although their longer-term outcomes and supportive care service needs have not been comprehensively documented. The aim of this study was to identify longer-term outcomes and supportive care needs in disease-free breast cancer survivors.
Materials and methods
One hundred seventeen patients who had been diagnosed with breast cancer 2–10 years earlier completed questionnaires to assess psychosocial outcomes including supportive care needs, psychological distress, and quality of life (QoL).
QoL and depression scores were consistent with community rates although anxiety scores were higher. Approximately two thirds of survivors reported at least one unmet need, most frequently concerning existential survivorship issues, thereby highlighting the unique needs of survivors. Years since diagnosis was not correlated with need levels. Survivors classified as clinically anxious reported over three times as many unmet needs and survivors classified as depressed reported over two and a half times as many unmet needs. Positive outcomes were frequently reported.
The findings have direct clinical relevance: irrespective of years since diagnosis, comprehensive and extended supportive care services are required to identify breast cancer survivors in need of supportive care interventions and remediate high levels of anxiety.
KeywordsBreast cancer survivorsUnmet needsSupportive carePsychosocial outcomesBenefit finding
Breast cancer is the most common registrable cancer in Australian women and affects one in 11 women under the age of 75 years . Five-year age-adjusted relative survival rates are over 80%  and women who have survived breast cancer constitute the largest population of cancer survivors . Survival rates are expected to continue to improve, resulting in an ever increasing population of individuals in our community who have survived cancer. Supportive care services involve the provision of services to meet physical, psychosocial, informational, practical, and spiritual needs experienced during the diagnostic, treatment, and follow-up phases of cancer . Supportive care efforts need to extend across the survivorship continuum: identifying and developing interventions to address the concerns and needs of this growing population of individuals has been identified as supportive care’s new challenge .
Definitions of “cancer survivors” vary and may include patients at any point from diagnosis, patients living with ongoing disease, and individuals who have been disease-free for five or more years . Little et al.  state: “The word ‘survival’ can mean different things to those who have cancer, those who care for them and those who treat their illness” (p 501). Mullan  provides a psychosocial definition of survival, which is helpful when examining adjustment responses over the time. Mullan describes survival as a progression of events definable in general outline: acute survival encompasses the diagnostic and treatment phase (which tends to occur during the first year), extended survival describes the period when active treatment has ceased and there is the greatest risk of recurrence for most cancers (from the end of the first year to 3 years), and permanent survival describes the subsequent period when the risk of recurrence for most cancers diminishes and patterns of previous lifestyle are re-established. In medical terms, women in the acute and extended survival phases may not yet be considered survivors due to the significant risk of disease recurrence, especially for those with advanced stage disease at diagnosis. The completion of treatment often heralds an expectation of a return to “normal life”, although longer survivorship periods do not necessarily confer improved psychosocial outcomes [6, 26, 28, 53]. Most studies of adjustment to breast cancer have focused on the acute and extended survival phases, and rates of psychosocial distress range between 20% and 43% within the first 1 to 2 years [16, 25, 31]. Women report transient and/or persistent physical, psychological, sexual, social, occupational and relationship problems [13, 32]. Elevated distress was associated with a number of variables including age, relationships status, psychiatric history, coping style, social support, self-esteem, menopausal status, education level, socioeconomic status, disease and treatment characteristics, and partner’s level of distress [10, 24, 34, 49]. All survivors live with the ongoing risk of disease recurrence, late effects and death [40, 41] and this chronic uncertainty is associated with elevated fears of recurrence [17, 22, 44].
A number of studies have examined the psychosocial outcomes of women two or more years post breast cancer diagnosis and despite methodological differences between studies , quality of life (QoL) is typically found to be within or above normal levels [12, 14, 20]. However, a significant minority of survivors experience a range of ongoing difficulties including depression, anxiety, body image disturbance, social problems, cognitive impairment, sexual difficulties, menopausal symptoms, family dysfunction, fatigue, pain, lymphedema, arm problems, insurance and work problems [12, 20, 23, 51, 57, 63]. Subgroups of survivors appear particularly vulnerable to specific difficulties [35, 42]; for example, women who have received chemotherapy report greater sexual dysfunction . Studies utilizing repeated assessments have found evidence of reduced distress over the long-term [21, 50], however, other studies have actually found higher distress five or more years post treatment than at earlier phases of the survivorship trajectory . Existential tensions  and changes in outlook and meaning in life  were reported. Many breast cancer patients perceive benefits after their experience [6, 9, 61, 66] and benefit finding has been correlated with lower socioeconomic status and advanced stage of disease , although findings are contradictory regarding the relationship with distress and QoL over the survivorship continuum [9, 61]. The assessment of such outcomes provides a more balanced understanding of survivorship and provides opportunities for improving adjustment, particularly when the amelioration of adverse effects is limited.
The presence of specific difficulties in a significant proportion of women apparently cured of breast cancer many years earlier focuses attention on the extended and permanent survival phases of their recovery [20, 46]. Gotay and Muraoka  reviewed research in this area and concluded: “Additional data are needed to understand the needs of long-term survivors... and to determine what kinds of support survivors want” (p 656). Need assessment measures provide a method to directly assess the gap between a person’s experience of services and the actual services required . Measures can identify high risk groups likely to benefit from targeted preventive and remediation interventions, and can facilitate the optimal allocation and delivery of limited health care resources . Level I evidence supports the efficacy of psychosocial interventions in adults with cancer [11, 45] and current best practice guidelines recommend the provision of psychosocial support to women diagnosed with breast cancer . However, understanding the nature and range of survivors’ actual supportive care service needs is critical if effective interventions are to be provided to improve outcomes.
Numerous measures have been developed to assess both the general needs and information needs of cancer patients [5, 29, 55, 56], carers , husbands  and their families . The psychometric status of many of these measures was not sufficiently evaluated [29, 55], although findings suggest that the needs of many patients are not being met [22, 64]. Higher levels of unmet needs have been identified in patients who were younger, female, unmarried, had advanced stage of disease at diagnosis, reported a psychiatric history and/or current psychological problems, poorer QoL, physical morbidity, lower income, financial difficulties and who live in a rural location, although unmet needs were not related to time since diagnosis or treatments received [3, 29, 54–56, 58, 62]. Thewes et al.  constructed a measure to assess the specific needs of early stage breast cancer survivors 6–24 months post cancer diagnosis. Elevated needs were identified in the information, medicalcommunication and psychological domains. Needs differed according to age with older survivors reporting higher information and medical communication needs, and younger survivors reporting higher psychological needs . The validity of Thewes’ measure in survivors beyond 2 years post diagnosis was not established and comparisons with survivors of other cancer sites is not possible. We recently developed a generic measure of supportive care needs and positive outcomes for cancer survivors who are at least 1 year post diagnosis and disease-free . Thus, it is now possible to quantitatively document and compare the supportive care needs of different populations of cancer survivors over the extended and permanent survival phases.
This study aims to identify the prevalence and correlates of supportive care needs in breast cancer survivors. The present study defined a “breast cancer survivor” as a woman who had a diagnosis of cancer at least 2 years earlier and is currently disease-free. Survivors who were up to 10 years post diagnosis were recruited into the study to assess outcomes during the extended and permanent survival phases.
Materials and methods
Consecutive patients treated for breast cancer in an outpatient radiation oncology department in a Sydney hospital were invited by letter from their oncologist to participate in the study. Patients were eligible if they had been seen for clinical follow-up within the previous 2 years, were currently disease-free, able to communicate in English, and free from major intellectual or psychiatric impairment. Patients who indicated interest in the study were telephoned by a research assistant who provided further verbal information and obtained verbal consent. Patients in a relationship were invited to ask their partner to participate in the study. Data from partners will be presented in a separate manuscript. Participants were mailed a participant information sheet and self-report questionnaire, and gave written consent in accordance with ethical protocols. The study procedure was approved by the relevant human research ethics committees.
A self-report questionnaire was constructed to assess basic demographic and disease related information.
QoL and psychological distress
QoL was assessed using the Medical Outcomes Study Health Survey Short Form-12 (SF-12) . The SF-12 is a 12-item measure that assesses both physical and mental QoL. The SF-12 has demonstrated validity and reliability, and normative data is available from the general USA population . Psychological distress was assessed using the Hospital Anxiety and Depression Scale (HADS) . The 14-item HADS has two subscales measuring anxiety and depression and has been used widely in community and patient populations and has proven validity and reliability . People scoring above 11 on each subscale are classified as a depression or anxiety case, while people scoring 8–10 are classified as a probable case.
Supportive care needs
Survivors’ supportive care needs were assessed using the CaSUN . The CaSUN contains 35 need items, 6 positive change items and 1 open-ended question. Need items are scored no need (0), met need (1) or unmet need (2); total scores are achieved by summing responses in that category. Other items are interpreted descriptively. The Cronbach alpha score for the CaSUN was 0.96, indicating a high degree of internal consistency. Construct validity was supported by a number of significant correlations between survivor variables and needs. Evidence for the face and content validity of the CaSUN was provided from responses on an evaluation sheet, and feedback supported the relevance, comprehensiveness and acceptability of the measure. Reading ease was satisfactory and the CaSUN takes approximately 10 min to complete. Maximum likelihood factor analysis identified five distinct need domains: (1) existential survivorship (e.g. cope with changes to my beliefs, make decisions about my life in context of uncertainty, explore spiritual beliefs); (2) comprehensive cancer care (e.g. doctors talk to each other to coordinate my care, manage health care with team, local health care services); (3) information (e.g. up to date information, understandable information); (4) quality of life (e.g. changes to quality of life, manage side effects); and (5) relationships (e.g. impact on my relationship, problems with my sex life). Cronbach alpha scores for the five factors ranged between 0.78 and 0.93, test–retest reliability for three of the five factors was good although was low overall, and this awaits further investigation.
The Statistical Package for the Social Sciences version 10.0 (SPSS for Windows) was used to analyse the data. Pearson product moment correlations with Bonferroni probabilities were conducted to examine relationships between survivor variables and met, unmet and total needs. One-way analysis of variance (ANOVA) tests were conducted to assess differences between groups of participants.
One hundred seventy-four breast cancer survivors were invited to participate in the study and 117 survivors returned completed questionnaires (67% response rate). Thirty-four survivors did not respond to the invitation and 23 refused to participate on the basis of insufficient English, the belief that they had nothing to offer the research, or a preference not to be reminded of the cancer.
Demographics and disease characteristics of breast cancer survivors (N = 117)
Year 10 or below
Previous cancer treatmenta
Mental QoL (mean = 52.3, SD = 9.5) and physical QoL (mean = 48, SD = 9.9) were within USA population norms (mean = 50, SD = 100) . Less than 1 third of survivors (28%) reported a history of mental health problems and a quarter (25%) had received psychological or psychiatric treatment. Six survivors (5.1%) reported symptoms of depression which met the diagnostic criteria for caseness (11+) on the HADS , and 11 survivors (9.4%) reported symptoms of anxiety that met the diagnostic criteria for caseness. Five survivors (4.3%) met the criteria for a probable case (8–10) of depression on the HADS and ten survivors (8.5%) met the criteria for a probable case of anxiety.
What are breast cancer survivors’ supportive care needs?
Breast cancer survivors’ endorsement of the CaSUN ranked by met need
Total met needs
Total unmet needs
Comprehensive cancer care
Quality of life
Breast cancer survivors’ most frequently endorsed met needs
CaSUN need item
I need to feel like I am managing my health together with the medical team
I need the very best medical care
I need to know that all the doctors talk to each other to coordinate my care
I need local health cancer services that are available
I need up to date information
I need information provided in a way I can understand
I need any complaints regarding my care to be properly addressed
I need help to manage my concerns about the cancer coming back
I need help to reduce stress in life
My family/partner needs information relevant to them/I need to talk to others who have experienced cancer
Breast cancer survivors’ most frequently endorsed unmet needs
CaSUN need item
I need help to manage my concerns about the cancer coming back
I need up to date information
I need information provided in a way that I can understand
I need an ongoing case manager to whom I can go to find out about services whenever they are needed
I need access to complementary and/or alternative therapy services
I need more accessible hospital parking
I need help to deal with my own and/or others expectations of me as a ‘cancer survivor’
Due to my cancer, I need help getting life and/or travel insurance
I need to know that all my doctors talk to each other to coordinate my care
I need help to cope with changes to my belief that nothing bad will ever happen in my life
Which breast cancer survivors report supportive care needs?
Survivors who met the HADS criteria for anxiety caseness had on average 14.2 unmet needs, compared to 6 unmet needs in survivors with probable anxiety, and 3.9 unmet needs in survivors with no anxiety. Clinically anxious survivors reported over three times (3.4) as many unmet needs as those with no anxiety, and there was a highly significant difference between survivors whose responses were classified as a no case, probable case or definite case of anxiety, F(2, 114) = 10.46, P < 0.001. In contrast, met needs were highest in survivors who did not meet the criteria for anxiety caseness (mean = 6.3), and lowest in survivors classified as anxious (mean = 4.7). Survivors who met the HADS criteria for depression caseness had on average 12.3 unmet needs, compared with 6.6 unmet needs in survivors with probable depression, and 4.6 unmet needs in survivors with no depression. Clinically depressed survivors therefore reported two and a half times (2.5) as many unmet needs as those with no depression. There was a significant difference between survivors whose responses were classified as a no case, probable case or definite case of depression, F(2, 114) = 3.13, P < 0.05. Again, met needs were highest in survivors who did not meet the criteria for depression caseness (mean = 6.1), and lowest for survivors classified as a definite case of depression (mean = 4.8).
Matrix of Bonferroni probabilities between survivor variables and the CaSUN
Total met needs (CaSUN)
Total unmet needs (CaSUN)
Number of treatments
Years since diagnosis
Physical QoL (QoL SF-12)
Mental QoL (QoL SF-12)
Do breast cancer survivors report positive outcomes?
Over three quarters (80%) of survivors reported at least one positive outcome and approximately half of the sample endorsed each positive change item. The most highly endorsed outcome, I have grown as a person was endorsed by 56% of survivors. Over half (55%) of survivors endorsed I focus more on things that are important, 51% endorsed I have made lots of positive changes in my life, 50% endorsed I appreciate my relationships with others more, 49% endorsed I realise how precious life is, and 46% endorsed I have benefited from contact with other cancer survivors and/or their families.
The study found that women with no evidence of current breast cancer reported ongoing anxiety and supportive care service needs 2–10 years after a breast cancer diagnosis, extending previous findings of distress and needs over the extended and permanent phases of survival [40, 64]. The absence of ongoing disease does not necessarily imply an absence of cancer related supportive care needs and longer time since diagnosis was not associated with lower needs. Breast cancer survivors did not report impaired QoL, which is consistent with other findings in breast cancer survivors [12, 23] and depression rates (5.1%) were consistent with age-adjusted community prevalence rates (5.3%) in Australia . However, anxiety rates (9.4%) were considerably higher than age-adjusted community prevalence rates (5.9%) in Australia . Anxiety rates were equivalent to those reported by women within the first 3 months of diagnosis (8.6%) and by women with advanced stage disease (6%) , indicating that anxiety continues to be a significant problem for many women years after the diagnosis of breast cancer. These findings support clinical guidelines recommending routine and ongoing psychosocial care .
Most survivors (86%) reported met supportive care needs. The domain of highest met need on the CaSUN was the comprehensive cancer care domain. Highly endorsed met needs related to health care access and coordination, and these more concrete needs may be better met by current cancer care services. Two thirds of survivors identified at least one unmet need and on average eight unmet needs. The domain of most frequently endorsed unmet needs was existential survivorship, which contains needs unique to survivors, such as making decisions in the context of uncertainty and existential issues. A number of survivor-specific unmet needs from the existential survivorship domain were highly endorsed, for example, dealing with expectations of self as a ‘cancer survivor’ and dealing with changes to my belief that nothing bad will happen in my life, reflecting the unique needs of survivor populations. The most prevalent unmet need, manage concerns about the cancer coming back was endorsed by one third of survivors. Fear of recurrence has been frequently identified in both qualitative and quantitative studies [17, 22, 44] and while this fear was found to diminish over the first 12 months , our findings indicate that significant numbers of survivors report persistent fears of recurrence up to 10 years post diagnosis.
One in four survivors reported an unmet need for up to date information, information provided in a way I can understand, an ongoing case manager and access to complementary and/or alternative therapy services, thereby providing clear guidance for service providers on service needs. High information needs are reported by patient populations [29, 54, 55, 62, 64] and reported in studies which have included both patients and survivors [19, 38]. For example, Girgis et al.’s  study of women with breast cancer found highest needs in the health/information and psychological domains, and Lintz et al.’s  study of men with prostate cancer found highest needs in the health system/information and psychological domains. Factor analysis of the CaSUN did not reveal a unique psychological domain, however, items relating to psychological needs were included in both the existential survivorship and relationships domains. High endorsement of needs from the existential survivorship domain is therefore consistent with high rates of psychological needs found in patient populations. In sum, our findings suggest that both information and psychological needs persist over the survivorship continuum and that new needs may arise, underscoring the importance of repeated assessment using appropriate and population-specific measures.
A number of variables have been associated with elevated needs in patients [19, 29, 54–56, 58, 62]: age, relationship status, employment status, psychological functioning and QoL were assessed in this study. Consistent with findings in patient populations, years since diagnosis was not related to needs, although physical QoL and employment status were also not found to be related to need levels. In our study, breast cancer survivors who were in a relationship and had received more treatments reported higher met needs. Survivors in a relationship may be better able to identify needs and access supportive care services, and partners may be able to provide support and facilitate access to services. Being in a relationship was associated with better mental and physical health outcomes and may buffer against the impact of stressful life events . Survivors who received more treatments may have increased contact with health care professionals, resulting in increased recognition of needs and provision of services. Lower mental QoL and higher depression and anxiety scores were associated with higher unmet needs and may represent vulnerability factors; for example, survivors who met caseness on the HADS for anxiety and depression demonstrated significantly higher levels of unmet needs than survivors who did not. Survivors classified as anxious reported over three times as many unmet needs and those classified as depressed reported over two and a half times as many unmet need. Met needs showed the opposite trend: higher met needs were not associated with caseness of depression or anxiety. These findings make clinical sense, however, it is unclear whether need levels are an antecedent or consequence of distress and these relationships await prospective investigation.
Over three quarters of breast cancer survivors endorsed at least one positive outcome, supporting and extending previous findings of multiple perceived benefits after cancer [6, 9, 61, 66]. Further investigation of correlates and predictors of benefits and therapeutic interventions to enhance benefit finding is warranted given the association with long-term positive effects in other populations of medical patients .
A number of limitations should be recognised. This cross-sectional study provides a snapshot of needs and a prospective study is currently underway to explore predictors of outcomes across the survivorship trajectory. Causal and contributing factors for the observed elevated anxiety rates in this sample cannot be assumed and need to be directly ascertained. Initial psychometric evaluation of the CaSUN found low test–retest reliability and we cannot assume the stability of needs identified over time . Thewes et al.’s  needs measure for breast cancer survivors may have provided more sensitive assessment of breast cancer specific issues although no additional needs were identified by more than one survivor in this study on the open-ended question. It was not used as it was not validated for use beyond 2 years post diagnosis and does not permit comparisons (which are planned) with other populations of survivors.
Although disease-free, over half of the survivors in this study were receiving prophylactic interventions (such as hormonal therapy) and many experience side effects as severe as those undergoing cancer treatment. These participants were retained in the study as their exclusion would have limited the generalizability of findings. In addition, while survivors were classified in psychosocial terms as being in the extended and permanent phases of survival, in medical terms, many of the women were in the very early phases of survival with regard to possible complications, secondary malignancies, and disease recurrence. The sample size was not considered sufficient to compare outcomes over extended and permanent phases of survivorship or conduct multiple regression analyses to identify predictors of outcomes. Age was found to influence needs in other survivor populations [5, 37, 59], but was not correlated with need levels in this study. The employment of previously used categorical definitions of age (e.g. 50 years or under, 51 years or over) resulted in only 14% of our sample being classified in the younger age group so was not explored further. Stage of disease data was unfortunately unavailable; distress and needs are likely to have differed according to disease stage  and further research could help identify whether women with more advanced disease are at risk of poorer outcomes and more in need of supportive care services.
Survivors were recruited from one urban Australian radiation oncology treatment centre and although patients attending the treatment centre were found to be comparable to other populations of breast cancer patients, our findings await replication. In addition, generalizability of the finding to other health care systems cannot be assumed. The assessment of needs within the dyadic relationship is merited given the established interdependence between patients’ and partners’ adjustment . Data from partners will be presented in a separate manuscript.
High rates of anxiety and supportive care needs are reported by breast cancer survivors up to 10 years post cancer treatment. QoL and depression rates were consistent with population rates and positive outcomes were frequently reported. These findings have direct clinical implications: all health care professionals involved in the follow-up care of breast cancer patients need to be aware that approximately one in ten disease-free survivors will experience clinical levels of anxiety, and survivors with high levels of distress are likely to have unmet needs for supportive care interventions, and vice versa. The size of the population of breast cancer survivors [3, 63] indicates that this presents a significant opportunity to assist a large number of women to achieve improved psychosocial outcomes. The CaSUN and HADS are useful for identifying such individuals. Many survivors will no longer attend follow-up clinics and innovative intervention strategies may be required, such as interventions delivered by primary care workers, community nurses, the Internet and self-help groups. Needs may not be effectively addressed by one-off interventions  and more systemic health care changes may be required. Other needs may prove “unmeetable” or beyond the scope of cancer services  and strategies to enhance positive outcomes may prove beneficial. This study indicates that comprehensive and extended follow-up supportive care services are required to better address anxiety and meet supportive care needs to improve the quality of breast cancer survivors’ adjustment over the extended and permanent phases of their recovery.
This research was supported by a scholarship from the NHMRC No: 358826 and a Training Fellowship in Psychiatric Research from the NSW Institute of Psychiatry, Australia. The research team would like to sincerely thank Ms. Denyse Stephens for her assistance with the data management.